Biosensors Based On Self-assembling Peptide Composites

In recent years,environmental pollution,food safety,clinical and industrial issues of people’s livelihood have promoted the rapid development of the field of detection.At present,the main problems of electrochemical sensors are the electron transfer rate of electroactive center and the retention and specificity of the electrocatalytic activity.Scholars at home and abroad usually choose more suitable substrate materials and electronic mediators to construct more diverse structures to solve these problems.Since self-assembling peptides have good biocompatibility and signal-conjugation properties,they have the potential to be used as substrate materials,and have become a research hotspot in the field of sensing in recent years.In this paper,a series of electrochemical biosensors are constructed based on self-assembling peptide composites to achieve efficient detection of substrates.The specific research content is divided into the following three parts:1.We design an amphiphilic self-assembling peptide Z23 with the following amino acid sequence:Ac-KFKFQFKFK-Am.The composite materials（Z23/MWCNTs）are formed by theπ-πstacking between the aromatic groups on the hydrophobic side of Z23 and the side walls of multi-walled carbon nanotubes（MWCNTs）.The positively charged Lys of the peptide make the complexes repel each other,improve the water solubility and enable the complexes to interact with target molecules.The sensors of nicotinamide adenine dinucleotide（NADH）and bisphenol A（BPA）were prepared by composite materials with good performance.During electrocatalysis process,the synergistic effect between Z23 and MWCNTs can significantly reduce overpotential of the component to be measured and increase response current.The linear range of NADH sensor is 1μM¹400μM,the sensitivity is 1.663μAμM^-1cm^-2.When the signal-to-noise ratio S/N is 3,the calculated detection limit is 0.34μM.The linear range of BPA detection is 1nM¹00μM with a sensitivity of 6.569μAμM^-1cm^-2,and detection limit of1.28nM.2.Prussian blue nanoparticles（PBNPs）are obtained by in-situ redox reaction based on self-assembling peptide Z23,followed by dispersion of carbon nanotubes to synthesize composite hydrogels（PBNPs/Z23/MWCNTs）.The composites are further placed on glassy carbon electrode to construct a biosensor for measuring hydrogen peroxide（H₂O₂）.During this process,Z23 not only prevents the agglomeration of nanoparticles,but also makes them more biocompatible.The performance parameters of H₂O₂ sensor are as follows:the sensitivity is203.7μAmM^-1cm^-2,response time is 3s,detection limit is 0.5μM,and the linear range is 1μM⁶00μM.In addition,the sensor has good stability and high selectivity in presence of other coexisting potential electroactive interfering substances.3.By combining electrochemical sensing and machine learning（ML）,intelligent data analysis are performed on the detection results.We use two established regression analysis algorithms,artificial neural network and support vector machine,to model the sample data of the current density of NADH and BPA with the concentration to be measured.Experimental results show that the prediction error of biosensor output is very low.Although some points are different from the observed target value,the difference is lower than the error of traditional linear regression.Consequently,ML can be used to analyze the output value of the sensor instead of the traditional regression.